docs: remove preliminary benchmarks pending official eval

fix: update links for merged branches, add 4bit quant scores (33.6%)

add hf-bf16/tokenizer_config.json from bf16-hf

add hf-bf16/tokenizer.json from bf16-hf

add hf-bf16/model.safetensors from bf16-hf

add hf-bf16/generation_config.json from bf16-hf

add hf-bf16/config.json from bf16-hf

Update README.md

Update README.md

docs: remove MLX from main, compact Use section with links

remove MLX model.safetensors — main is HF transformers + GGUF

add lemer-bf16.gguf from bf16-gguf

add lemer-q8_0.gguf from 8bit-gguf

add lemer-q6_k.gguf from 6bit-gguf

add lemer-q5_k_m.gguf from 5bit-gguf

add lemer-q3_k_m.gguf from 3bit-gguf

docs: add title attributes with run history to benchmark links

docs: merge best-per-category scores into grid, link to bf16 methodology

docs: remove best-per-category from main, keep raw grid only

docs: add MMLU-Pro best-per-category table (46.4%), 5-shot results

docs: add 5-shot CoT methodology note, mlx_lm verification

data: add Lemer(1,1) full 14-category results with transcripts

docs: fill Lemer(1,1) scores across all 14 categories

Update README.md

fix: config.json matches HF 4-bit safetensors (BitsAndBytes NF4)

add HF transformers 4-bit (NF4) safetensors to main

remove MLX model.safetensors.index.json from main

remove MLX model.safetensors from main

fix: use full base config with quantisation field for correct param count

add processor_config.json from base model

Update README.md

Update README.md

add 4bit GGUF: lemer-q4_k_m.gguf

add 4bit MLX: tokenizer_config.json

add 4bit MLX: tokenizer.json

add 4bit MLX: model.safetensors.index.json

add 4bit MLX: model.safetensors

add 4bit MLX: generation_config.json

add 4bit MLX: config.json

add 4bit MLX: chat_template.jinja

docs: remove MMLU-Pro from training datasets (benchmark only)

docs: add LEM-research dataset reference

docs: link quant table headers to branches

data: add benchmark result stock-e2b-bf16-math-think-temp1.json

data: add benchmark result stock-e2b-bf16-math-temp0.json

data: add benchmark result stock-e2b-bf16-math-nothink-temp1.json

data: add benchmark result stock-e2b-bf16-math-nothink-temp0.json

data: add benchmark result stock-e2b-bf16-biology-temp0.json

data: add benchmark result lemer-bf16-math-think-temp1.json

data: add benchmark result lemer-bf16-math-temp0.json